TDL: added kwargs_grid_search parameter, which is passed to the GridSearchCV class. Gives the user more freedom in setting up the GridSearch. For cv, which was manually set before the standart value is set to 10 s.t. previous behavior is preserved. (resolves #546)

Author: cwork

ontolearn/learners/tree_learner.py

@@ -157,6 +157,7 @@ def __init__(self, knowledge_base,
  max_runtime: int = 1,
  grid_search_over: dict = None,
  grid_search_apply: bool = False,
+ kwargs_grid_search: dict = None,
  report_classification: bool = True,
  plot_tree: bool = False,
  plot_embeddings: bool = False,
@@ -182,13 +183,17 @@ def __init__(self, knowledge_base,
  pass
  else:
  grid_search_over = dict()
+
+ kwargs_grid_search.setdefault("cv", 10)
+ 
  assert (
  isinstance(knowledge_base, KnowledgeBase)
  or isinstance(knowledge_base, ontolearn.triple_store.TripleStore)
  or isinstance(knowledge_base)
  ), "knowledge_base must be a KnowledgeBase instance"
  print(f"Knowledge Base: {knowledge_base}")
  self.grid_search_over = grid_search_over
+ self.kwargs_grid_search = kwargs_grid_search
  self.knowledge_base = knowledge_base
  self.report_classification = report_classification
  self.plot_tree = plot_tree
@@ -387,7 +392,7 @@ def fit(self, learning_problem: PosNegLPStandard = None, max_runtime: int = None
  if self.grid_search_over:
  grid_search = sklearn.model_selection.GridSearchCV(
  tree.DecisionTreeClassifier(**self.kwargs_classifier),
- param_grid=self.grid_search_over, cv=10, ).fit(X.values, y.values)
+ param_grid=self.grid_search_over, **self.kwargs_grid_search).fit(X.values, y.values)
  print(grid_search.best_params_)
  self.kwargs_classifier.update(grid_search.best_params_)
  # Training